Journal of Medicinal Chemistry, 1972, Vol. 1.5, No. 8
Folate Antagonists. 3
827
Folate Antagonists. 3. 2,4-Diamino-6-(heterocyclic)quinazolines,a Novel Class of Antimetabolites with Potent Antimalarial and Antibacterial Edward F. Elslager,* J. Clarke, Leslie M. Werbel, Donald F. Worth, Department of Chemistry, Research and Development Division, Parke, Davis and Company, Ann Arbor, Michigan 481 06
and John Davoll Department of Chemistry, Research and Development Division, Parke, Davis and Company, Hounslow, Middlesex, England. Received February 23, 19 72
Eleven 2,4-diamino-6-(heterocyclic)quinazolines(VI) were prepared by condensing 5-chloro-2-nitrobenzonitrile (111) with the appropriate saturated heterocycle to give the corresponding 5-(heterocyclic)-2nitrobenzonitriles (IV) (24-94%), followed by SnClz. 2Hz0 reduction to the 2-amino-S-(heterocyclic)benzonitriles (V) (5 1-85%) and cyclization to VI (6-7 1%) with chloroformamidine SHC1 or cyanoguanidine. The reaction of 2,4;6;triaminoquinazoline (33) with 2,5-diethoxytetrahydrofuran,acetonylacetone, phthalic anhydride, and c i , -dibromoxylene ~ afforded 2,4-diamino-6-(pyrrol- 1-y1)quinazoline (34), 2,4(35), N-( 2,4-diamino-6-quinazolinyl)phthalimide(36), diamino-6-( 2,5-dimethylpyrrol-1-yl)quinazoline and 2,4-diamino-6-( 2-isoindoliny1)quinazoline (37) in 26-8 5% yield. 2,4-Diamino-6-(3,5-dimethylpyrazol(39) by cyclization 1-y1)quinazoline (40) (63%) was prepared from 2,4-diamino-6-hydrazinoquinazoline with acetylacetone. Ring closure of 2,4,6,8-tetraaminoquinazoline(43) and 2,4-diaminod-(aminomethyl)quinazoline (46) utilizing acetonylacetone gave 2,4-diamino-6,8-bis( 2,5-dimethylpyrrol-l-yl)quinazoline (44) (53%) and 2,4-diamino-6-[(2,5-dimethylpyrrol-l-yl)methyll quinazoline (47) (69%), respectively. Thirteen heterocyclic quinazolines were active orally or parenterally against Plasmodium berghei in mice and P. gallinaceum in chicks. 2,4-Diamino-6-(2-phenyl-l-pyrrolidinyl)quinazoline (27) and 2,4-diamino-6(2-benzy1piperidino)quinazoline (3 1) showed activity comparable with or superior to cycloguanil hydrochloride and 2,4-diamino-6-[(3,4-dichlorobenzyl)nitrosamino]quinazoline(Ib). Many of the heterocyclic quinazolines were also highly active against a broad spectrum of pathogenic and nonpathogenic bacteria in vitro. 2,4-Diamino-6-(2,5-dimethylpyrrol1-yl)quinazoline (35) exhibited good activity against Streptococcus p y o g e n e s (C-203) in mice following single oral or sc doses ranging from 50 to 500 mg/kg, and the drug was synergistic with sulfamethoxypyridazine. In recent communications from these laboratories relating to folate antagonists,'-' it was reported that various 2,4-diamino-6- { [aralkyl and (heterocyclic)methyl] amino- and -nitrosamino)quinazolines, exemplified by 2,4-diamino-6[(3,4-dichlorobenzyl)amino]quinazoline (Ia)'y3 and 2,4-diI1
b;R=NO
amino-6- [(3,4-dichlorobenzyl)nitrosamino]quinazoline (Ib),46 exhibit strong antimalarial activity against sensitive and drugresistant lines of Plasmodium berghei in mice, P. gallinaceum in chicks, and P. cynomolgi and P. knowlesi in rhesus These substances also possess an encouraging degree of activity against Trypanosoma cruzi in chick embryo cell cultures and in m i ~ e ' 9 ~ 3and ~ 4 are potent folate antagonists in bacterial In contradistinction, various 2,4diamino- and 2-amino4hydroxyquinazoline Glu and Asp analogs7 (11, where x = 1 or 2; R = H or CH3; and X = OH or NH2) lack appreciable antiparasitic activity, but display potent inhibitory effects against Streptococcus faecalis R (ATCC 8043)' [Strep. faecium var. durans (SF/O)] ,9 thymidylate synthetase and dihydrofolic reductase from mammalian and bacterial sources,8 several tThis is paper 25 of a series on antimalarial drugs. For paper 24, see ref 1 . $This investigation was supported in part by U. S. Army Medical Research and Development Command Contract DA-49-193-MD2754. This is Contribution No. 922 to the Army Research Rogram on Malaria.
sublines of the L1210 mouse leukemia,'and L1210 cells in ~ u l t u r e . ~ We now report the synthesis and biological properties of various 2,4-diamino-6-(heterocyclic)quinazolines(VI), a novel series of antimetabolites with potent antimalarial and antibacterial activity. Chemistry. Many of the 2,4-diamino-6-(heterocyclic)quinazolines (VI) were prepared utilizing the route exemplified by Scheme I. The condensation of 5-chloro-2-nitroScheme I
CI I H,NC=NH, HCI NH /I or H,NCNHCN 6-7 1%
V
VI
benzonitrile (111) with the appropriate saturated heterocycle in 2-ethoxyethanol or DMSO afforded the corresponding 5-(heterocyclic)-2-nitrobenzonitriles(IV, 1-10, Table I) in 26-94%yield (procedures I, 11). Reduction of the nitro-
828
Elslager, et al.
Journal ofMedicina1 Chemistry, 1972, Vol. IS,No. 8
Table I. 5-(Heterocyclic)-2-nitrobenzonitriles
No.
l
2
MP, "C
Purificn solvent
Procedure
Formula
Analyses
127-128
94
EtOH
i
161-163
85
EtOH
1
180-1 82
76
MeCN
1
67
DMSO
11
Ci 7H13CLN3Oz
C, H,
189-192
90
MeCN
11
1'
C, H, N
188-191
77
DMSO
11
C17H15N302
162-164
39
EtOH
186-1 87
26
EtOH
1
C l , H l , N 3 0 ~ ~ 0 . 1 H z 0 C, H, N, HZO
H 3 C 6 H 5
220-222
86
DMF-H,O
1
C18Hl TN3'2
C, H,
o C H 2 C 6 H 5
120-126
47
EtOH
11
C19H19N302
C, H, N
n
N
NCH,
u
4
5
W
6
rkca5
7
Yield purified, %
W
C
C
I
l
3N302
CL2H14N402
C, H. N C, H,
N
C. H, N
,H1,C1N302
N
C, H, N
n
H3C 9
10
N
&.qNH2 &gqNHZ
Scheme I1
0
34
CH3 35 1
benzonitriles (IV) with SnCl2-2H2Oin aqueous HC1 or aqueous HCl-HOAc gave the 2-amino-5-(heterocyclic)benzonitriles (V, 11-20, Table 11) in 5 1-85% yield (procedures
36 0
111-V). Cyclization of the requisite aminobenzonitrile (V) with chloroformamidine hydrochloride'o in dry diglyme (procedure VI) or with cyanoguanidine in concd HC1 (pro-
Journal of Medicinal Chemistry, 1972, Vol. 15, No. 8
Folate A n tag0 nists. 3
829
Table 11. 2-Amino-5-(heterocyclic)benzonitriles
No.
Yield purified, %
Purificn solvent
Procedure
92-94
12
CC14-petr ether
111
CIZHIIN~
C, H, N
98-100
84
cc1,
IV
C12H16N4
C, H, N
117-120
58
EtOH-H,O
V
C14H19N3
C, H, N
128-130
85
CROH
V
C16H15N3
134-1 36
56
CROH
V
1‘
O
11
N
12
NuNCH,
13
MP, C 3
n N
a
Formula
Analyses
‘lH5
15
SaZN3
C, H, N
16
W C l
135-137
51
i-ROH
V
Ci~Hie.aN3
C, H, N
17
U C 6 H 5
217-221
68
EtOH-Et,O
V
CiWi,N,.HCl
C, H, Cl-,N
143-144
61
EtOH-H,O
V
CisHi9N3.0.06HiO
C, H, N, HZO
m 18
NJC6H5 H3C
19
N3C,H,
166-167
79
i-ROH
V
C18H19N3
C, H, N
20
oCH,C6H5
104-105
15
EtOH-H,O
V
C19H21N3
C, H, N
cedure VII) gave the desired 2,4-diamino-6-(heterocyclic)quinazolines [V,21-31,Table 111) (6-71%). Alternatively, various 2,4-diamino-6-(heterocyclic)quinazolines were synthesized from 2,4,6-triaminoquinazoline (33)7 via 2,4-diamino-6-nitroquinazoline (3217(Scheme 11). Thus the condensation of 33 with 2,5-diethoxytetrahydrofuran, acetonylacetone, phthalic anhydride, and cup’-dibromo-o-xylene afforded 2,4-diamino-6-(pyrrol- 1 -yl)quinazoline (34)(26, 30%), 2,4-diamino-6-(2,5-dimethylpyrrol1-y1)quinazoline (35) (40,48%), N-(2,4-diamino-6-quinazoliny1)phthalimide (36)(85%), and 2,4-diamino-6-(2-isoindoliny1)quinazoline (37)(75%), respectively. Diazotization of 33 gave the diazonium salt 38,which was reduced in situ with SnC12.2H20 to give 2,4-diamino-6-hydrazinoquinazoline (39)(42,54%). Cyclization of 39 with acetylacetone afforded 2,4-diamino-6-(3,Sdimethylpyrazol-l-yl)quinazoline (40)(26,63%). 2,4-Diamino-6,8-bis(2,5-dimethylpyrrol1-yl)quinazoline (44)was readily synthesized starting from 3,S-dinitro-oanisonitrile (41)” (Scheme 111). Ring closure of 41 with guanidine hydrochloride gave 2,4-dimino-6,8-dinitroquinazoline (42)in high yield (85,86%). Hydrogenation of 42 over 10% Pd/C afforded 2,4,6,8-tetraahinoquinazoline(43) (93,96%), which was condensed with 2 equiv of acetonylacetone to give 2,4-diaminod,8-bis(2,5-dimethylpyrrol1 -yl> quinazoline (44)(30, 53%). 2,4-Diamino-6- [(2,5-dimethylpyrrol-l-yl)methyl] quinazoline (47),the methylene homolog of 35,was also prepared starting from 2,4,6-triaminoquinazoline(33)7 (Scheme IV).
Scheme 111
NH, 42
41 NH,
43 H,C
J n N
CH,
@TN% &H
NH, CH,
44
The triamine was first converted via the Sandmeyer reaction to the known 2,4diamino-6quinazolinecarbonitrile (45),7 which was hydrogenated over Raney Ni at 70-75’ (1 00 atm) to give 2,4-diamino-6-(aminomethyl)quinazoline(46)in 5382% yield. Cyclization of 46 with acetonylacetone afforded the desired product (47)in 60459% yield. Spectral data (ir, uv, nmr) were in agreement with the structures shown for each of the 2,4-diaminod-(heterocyclic)quinazolines.
830
Elslager, et al.
Journal of Medicinal Chemistry, 1972, Vol. 15, No. 8
Table 111. 2,4-Diamino-6-(heterocyclic)quinazolinesvia 5-Chloro-2-nitrobenzonitrile
No.
MP, "C
21
n
N
22
NCH,
u
N3
23
Yield purified, %
292-294
40
286-292
71
320-323
29
Na
Purificn solvent
EtOH-H,O
EtOH-H,O
Procedure
VI
Formula
Analyses
7N 5
C, H, N
VI
Ci3HisN6
C, H, N
VI
C, ,H,,N,. HC1. 0.05H20
C, H, Cl-, N,
1
'ZH5
h20
248-25 1
34
MeOH
v11
C,7H,7N5~2C7H803SaC, H, N, S
25
149-152
31
i-&OH-MeCN
VI
cl IH1 7C12N5
C, H, N
26
145-147
45
i-Pr OH-MeCN
VI
27
246-241
26
EtOH
v11
28
200-205 dec
21
i-PrOH-MeCN
VI
C19H20C1N5
H, N; Cb
44
DMF
VI
C1,H,,N5~HC1
C, H, Cl-. N
28
MeOH
v11
C19H21N5
C, H, N
6
EtOH
VI
C2,H,,N5~0.25H,0
C, H, N,H,O
24
>300
29
30
255-257 dec
31
108-112
'T7H,0,S represents p-toluenesulfonic acid. k : calcd, 64.5; found, 63.9. Scheme IV
33
45
Ni, H, 5 3 , 8 0 , 82%.
W N H (CH&OCH,),, 2
60,69%
H,NCH, NH, 46
47
The pyrrolidine, piperidine, and piperazine intermediates required for the synthesis of compound 21-24,30, and 31 are commercially available. 2-(3,4-Dichlorophenyl)pyrrolidine, 2-@-chlorophenyl)pyrrolidine, l2 and 2-phenylpyrrolidine13 were obtained by the reaction of the requisite phenylmagnesium bromide and 4-chlorobutyronitrile to give the corresponding 2-phenyl-l-pyrroline~,'~ which were hydrogenated over 5% Pt/C. 2-@-Chloropheny1)piperidine was pre-
pared by the condensation of p-chlorophenylmagnesium bromide with 5-chlorovaleronitrile to give 2-@-chlorophenyl)3,4,5,6-tetrah~dropyridine,'~ followed by hydrogenation in HOAc over Pt/C. Attempts to reduce 2-methyl-3-phenyl-Ipyrroline to 2-methyl-3-phenylpyrrolidine with NaBH4 were unsuccessful, but catalytic hydrogeqation utilizing Pt/C afforded the intermediate 2-methyl-3-phenylpyrrolidjne. Antimalarial Effects. Antimalarial studies with the 2,4diamino-6-(heterocyclic)quinazolinesdescribed in the present communication were carried out utilizing P.berghei in mice and P. gallinaceum in chicks. Compounds 23,25,26,28,29, and 31 were administered sc in a single dose to mice infected withP. bergheis,#(Table IV). Five substances (23; 25,26, 28,31) cured all of the mice at one or more dose levels ranging from 40 to 640 mg/kg, but four of them, like cycloguanil hydrochloride, were toxic for mice at the higher dose levels. 2,4-Diamin0-6-(2-benzylpiperidino)quinazoline(3 l), the most promising member of the series, exhibited activity comparable with or superior to cycloguanil hydrochloride, 2,4-diamino-6- [(3,4-dichlorobenzyl)amino] quinazoline (Ia), and 2 ,Cdiaminod- [(3,4-dichlorobenzyl)nitrosamino]quinazoline (Ib). $The parenteral antimalarial screening was carried out by Dr. Leo Rane of the University of Miami, and test results were provided through the courtesy of Dr. David P. Jacobus, Dr. T. R . Sweeney, and Dr. E. A. Steck of the Walter Reed Army Institute of Research. #For a description of the test method, see ref 1 5 .
Journal of Medicinal Chemistry, 1972, Vol. 15, No. 8
Folate Antagonists. 3
831
Table IV. Parenteral Effects of 2,4-Diamino-6-(heterocyclic)quinazolines against Plasmodium berghei in Mice
AMST; T or C? after single mg/kg dose
NO.^
640
23
T5
25
T5
26
C2, T3
28
29 31
1.9 c5
320
160
80
40
20
10
c5 c5 C2, T3 C2, T3 C2, T3 C2, T3 T5
c5 c5 C2, T3 C2, T3 C4, T1 C4, T1 C2, T3 C1, T4 0.5 c5 c5
13.7; C1 13.4; C1 c5 c5 c5 c5 c5 c5 0.3 c5 c5
8.9 8.7 9.9; c 4 20.4; C3 20.9; C3 17.9; C4 c5 c5 0.3 c5 c5 c5 13.4; C1 13.4; C1 7.1 7.3
3.3 3.1 7.5 7.7 18.2; C2 21.1; c1 1.5 7.3 0.3 9.9; c 3 11.4; C3 12.4; C3 7.9 8.1 2.5 2.1 22.4; C1
0.7
0.5 c5 c5
2.5 7.7 4.3b 4.5 6.7c 6.9
4.9 C3, T2 21.6; C2 c5 C2, T3 21.9; C2 c5 12.9 0.7 Ia acetate c5 c5 9.9; c 3 13.1 0.7 c5 9.9; c 3 Ib acetate c5 c5 =AMST is themean survival time (days) of treated mice (MSTT) minus the mean survival time (days) of control mice (MSTC). In the present study the MSTC was 6.1 days. T signifies the number of toxic deaths, occurring on days 2-5 after infection, which are attributed to drug action. C indicated the number of mice surviving at 60 days post infection and termed "cured"; data to establish parasitological cure based on subinoculation are unavailable. Each compound was administered as a single sc dose. Each entry at each dose level represents results with a 5 animal group. bAMST at 5 mg/kg = 2.7 days; CAMST = 5.1 days at 5 mg/kg, 1.7 days at 2.5 mg/kg, and 0.5 days at 1.25 mg/kg. dStructures for het N are given in Table III. Cycloguanil hydrochloride
T5
Twelve compounds (21,22,24,27,30,34-37,40,44,47) were given by gavage for 4 days or continuously b y drug-diet for 6 days to mice infected with another normal drug-sensitive strain of P. berghei**ytt (Table V). Among them, seven quinazolines (21,24,27,30,34,44,47) produced a 90% suppression of the parasitemia at daily oral doses ranging from 0.35 to 16 mg/kg. 2,4-Diamino-6-(2-phenyl1-pyrrolidiny1)quinazoline (27),the outstanding member of the group, was approximately 210 times as potent as quinine hydrochloride and compared favorably with 2,4-diamino- [6(3,4-dichlorobenzyl)nitrosamino]quinazoline (Ib).*y6 Several of the heterocyclic quinazoline derivs (23,25,26) were also evaluated against P. gallinaceum infections in w h t e Leghorn cockerels.$$ Chicks were given an iv injection of 0.2 ml of heparinized blood infected with P. gallinaceum and having a minimum of 80-90% parasitized red blood cells. The parasitized blood was drawn by cardiac puncture from donor birds infected 72 hr earlier with P. gallinaceum. Donor strains were maintained in separate groups of chicks, 14-16 days old, that also received inoculations of heparinized infected blood. In every experiment 100% of the untreated control birds died within 72-96 hr postinfection. Candidate substances were administered to chicks in a single sc dose in peanut oil immediately after infection. In this test, as in the mouse test, the antimalarial activity of candidate compounds was assessed by comparing the maximum survival times of treated malaria-infected chicks with the survival times of untreated malaria-infected controls. A compd was arbitrarily **Selected compounds were kindly studied orally against P. berghei in mice by Dr. Paul E. Thompson and coworkers, Department of Pharmacology, Parke, Davis and Co., Ann Arbor, Mich. ??For a description of the test method, see ref 3 and 4. $$Antimalarial screening against Plasmodium gallinaceum in chicks was carried out by Dr. Leo Rane at the University of Miami, and test results were supplied through the courtesy of Dr. David P. Jacobus, Dr.T. R. Sweeney, and Dr.E. A. Steck of the Walter Reed ~
Army Institute of Research.
considered to be active against malaria if it produced increases in the survival times of treated chicks that were at least 100% over the survival times of untreated controls. As indicated in Table VI, all these substances were active based on these criteria and each prolonged the survival time of chicks for periods of 3.8-16.0 days following a single sc or iv dose ranging from 7.5 to 160 mg/kg. However, none was as promising as cycloguanil hydrochloride or the two diaminoquinazoline reference drugs (Table VI). Antibacterial Studies. Each of the 2,4-diamino-6-(heterocyc1ic)quinazolines was tested in vitro against a spectrum of pathogenic bacteria including Streptococcus faecalis (MGH-2), normal (UC-76) and drugresistant (S187 13) Staphylococcus aureus, Pseudomonas aeruginosa (28), Escherichia coli (Vogel), and Shigella sonnei ((2-10). A modification of the gradient plate procedure of Szybalski16 and Webb and Washington" was employed throughout. Fourteen compounds were active against Strep. faecalis (MGH-2) at
